Today, Transluce is releasing the most expansive independent evaluation to date of how AI systems respond to users experiencing mental health crises.
We evaluated 77 model variants from OpenAI, Anthropic, Google, Meta, SpaceXAI, Thinking Machines, DeepSeek and Moonshot AI.
Someone open sourced a repo that runs Claude Code for free
It hijacks your Claude Code traffic and points it at DeepSeek, Kimi and 8 other providers instead of Anthropic. same CLI, same workflow, none of the bill 5 minutes to set up. 36,913 devs already running it
DeepSeek v4 Flash Vision Exp has moved!
Image recognition was also confirmed. thank you! MiaAI_lab
There are more strong AI models than ever
But it’s hard to tell how differently they actually handle the same task until you put them side by side. So we tested Claude Opus 5, Kimi K3, GPT-5.6 Sol, and DeepSeek V4 Pro in ALwith, using the exact same prompt for each test:
You have reached the end of the archive
All of deepseek